Exploratory Analysis
#setwd('/afs/inf.ed.ac.uk/user/s17/s1725186/Documents/PhD-Models/FirstPUModel/RMarkdowns')
library(tidyverse) ; library(reshape2) ; library(glue) ; library(plotly) ; library(plotlyutils)
library(RColorBrewer) ; library(viridis) ; require(gridExtra) ; library(GGally)
library(Rtsne)
library(ClusterR)
library(DESeq2) ; library(biomaRt)
library(knitr)Load preprocessed dataset (preprocessing code in 20_04_02_data_preprocessing.Rmd)
# Gandal dataset
load('./../Data/preprocessed_data_imputed.RData')
datExpr = datExpr %>% data.frame
DE_info = DE_info %>% data.frame
# GO Neuronal annotations: regex 'neuron' in GO functional annotations and label the genes that make a match as neuronal
GO_annotations = read.csv('./../Data/genes_GO_annotations.csv')
GO_neuronal = GO_annotations %>% filter(grepl('neuron', go_term)) %>%
mutate('ID'=as.character(ensembl_gene_id)) %>%
dplyr::select(-ensembl_gene_id) %>% distinct(ID) %>%
mutate('Neuronal'=1)
# Update DE_info with Neuronal information
DE_info = DE_info %>% mutate('ID'=rownames(.)) %>% left_join(GO_neuronal, by='ID') %>%
mutate(Neuronal=ifelse(is.na(Neuronal), 0, Neuronal)) %>%
mutate(significant=padj<0.05 & !is.na(padj))
rm(GO_annotations)All samples together
There seem to be two different behaviours in mean expression by gene
A sort of heavy right tail in the samples’ mean expression, although the difference doesn’t seem to be that big when considering the scale of the x axis
plot_data = data.frame('ID'=rownames(datExpr), 'Mean'=rowMeans(datExpr))
p1 = ggplotly(plot_data %>% ggplot(aes(Mean)) + geom_density(color='#0099cc', fill='#0099cc', alpha=0.3) +
scale_x_log10() + theme_minimal())
plot_data = data.frame('ID'=colnames(datExpr), 'Mean'=colMeans(datExpr))
p2 = ggplotly(plot_data %>% ggplot(aes(Mean)) + geom_density(color='#0099cc', fill='#0099cc', alpha=0.3) +
theme_minimal() + ggtitle('Mean expression density by Gene (left) and by Sample (right)'))
subplot(p1, p2, nrows=1)rm(p1, p2, plot_data)Grouping genes by Neuronal tag and samples by Diagnosis
The two groups of genes seem to be partially characterised by genes with Neuronal function, but it doesn’t play such an important role as when we were considering also the non protein-coding genes.
The heavy right tale from the sample distribution corresponds to a group of Autism samples. In general, the autism group has a bigger mean and is more spread out than the control group
plot_data = data.frame('ID'=rownames(datExpr), 'Mean'=rowMeans(datExpr)) %>%
left_join(GO_neuronal, by='ID') %>% mutate('Neuronal'=ifelse(is.na(Neuronal),F,T))
p1 = plot_data %>% ggplot(aes(Mean, color=Neuronal, fill=Neuronal)) + geom_density(alpha=0.3) +
scale_x_log10() + theme_minimal() + theme(legend.position='bottom') +
ggtitle('Mean expression density by gene')
plot_data = data.frame('ID'=colnames(datExpr), 'Mean'=colMeans(datExpr)) %>%
mutate('ID'=substring(ID,2)) %>% left_join(datMeta, by=c('ID'='Dissected_Sample_ID'))
p2 = plot_data %>% ggplot(aes(Mean, color=Diagnosis, fill=Diagnosis)) + geom_density(alpha=0.3) +
theme_minimal() + theme(legend.position='bottom') +
ggtitle('Mean expression density by Sample')
grid.arrange(p1, p2, nrow=1)
rm(GO_annotations, plot_data, p1, p2)ASD vs CTL
In general there doesn’t seem to be a lot of variance in mean expression between autism and control samples by gene
plot_data = data.frame('ID'=rownames(datExpr),
'ASD'=rowMeans(datExpr[,datMeta$Diagnosis=='ASD']),
'CTL'=rowMeans(datExpr[,datMeta$Diagnosis!='ASD'])) %>%
mutate(diff=ASD-CTL, abs_diff = abs(ASD-CTL)) %>%
mutate(std_diff = (diff-mean(diff))/sd(diff), distance = abs((diff-mean(diff))/sd(diff)))
plot_data %>% ggplot(aes(ASD, CTL, color = distance)) + geom_point(alpha = plot_data$abs_diff) + geom_abline(color = 'gray') +
scale_color_viridis(direction = -1) + ggtitle('Mean expression ASD vs CTL') + theme_minimal() + coord_fixed()
summary(plot_data$std_diff)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -5.67843 -0.58822 -0.03302 0.00000 0.53143 7.81280cat(paste0('There are ', sum(plot_data$distance>3), ' genes with a difference between Diagnoses larger than 3 SD to ',
'the distance distribution of all genes'))## There are 202 genes with a difference between Diagnoses larger than 3 SD to the distance distribution of all genes#cat(paste0('Outlier genes: ', paste(plot_data$ID[abs(plot_data$std_diff)>3], collapse = ', ')))Grouping genes and samples by Diagnosis
There doesn’t seem to be a noticeable difference between mean expression by gene between Diagnosis groups
Samples with autism tend to have a wider dispersion of mean expression with higher values than the control group (as we had already seen above)
plot_data = rbind(data.frame('Mean'=rowMeans(datExpr[,datMeta$Diagnosis=='ASD']), 'Diagnosis'='ASD'),
data.frame('Mean'=rowMeans(datExpr[,datMeta$Diagnosis!='ASD']), 'Diagnosis'='CTL')) %>%
mutate('Diagnosis'=factor(Diagnosis, levels=c('CTL','ASD')))
p1 = ggplotly(plot_data %>% ggplot(aes(Mean, color=Diagnosis, fill=Diagnosis)) +
geom_density(alpha=0.3) + scale_x_log10() + theme_minimal())
plot_data = rbind(data.frame('Mean'=colMeans(datExpr[,datMeta$Diagnosis=='ASD']), 'Diagnosis'='ASD'),
data.frame('Mean'=colMeans(datExpr[,datMeta$Diagnosis!='ASD']), 'Diagnosis'='CTL')) %>%
mutate('Diagnosis'=factor(Diagnosis, levels=c('CTL','ASD')))
p2 = ggplotly(plot_data %>% ggplot(aes(Mean, color=Diagnosis, fill=Diagnosis)) +
geom_density(alpha=0.3) + theme_minimal() +
ggtitle('Mean expression by Gene (left) and by Sample (right) grouped by Diagnosis'))
subplot(p1, p2, nrows=1)rm(p1, p2, plot_data)Visualisations
Samples
PCA
The first principal component seems to separate perfectly the two diagnosis
pca = datExpr %>% t %>% prcomp
plot_data = data.frame('ID'=colnames(datExpr), 'PC1' = pca$x[,1], 'PC2' = pca$x[,2]) %>%
mutate('ID'=substring(ID,2)) %>% left_join(datMeta, by=c('ID'='Dissected_Sample_ID')) %>%
dplyr::select('ID','PC1','PC2','Diagnosis') %>%
mutate('Diagnosis'=factor(Diagnosis, levels=c('CTL','ASD')))
plot_data %>% ggplot(aes(PC1, PC2, color=Diagnosis)) + geom_point() + theme_minimal() + ggtitle('PCA') +
xlab(paste0('PC1 (',round(100*summary(pca)$importance[2,1],1),'%)')) +
ylab(paste0('PC2 (',round(100*summary(pca)$importance[2,2],1),'%)'))
rm(pca, plot_data)MDS
Looks exactly the same as the PCA visualisation, just inverting the x axis
d = datExpr %>% t %>% dist
fit = cmdscale(d, k=2)
plot_data = data.frame('ID'=colnames(datExpr), 'C1'=fit[,1], 'C2'=fit[,2]) %>%
mutate('ID'=substring(ID,2)) %>% left_join(datMeta, by=c('ID'='Dissected_Sample_ID')) %>%
dplyr::select('C1','C2','Diagnosis') %>%
mutate('Diagnosis'=factor(Diagnosis, levels=c('CTL','ASD')))
plot_data %>% ggplot(aes(C1, C2, color=Diagnosis)) + geom_point() + theme_minimal() + ggtitle('MDS')
rm(d, fit, plot_data)t-SNE
All of the results separate perfectly the samples by diagnosis, perhaps the one with the parameter perplexity = 10 separates them best. The results obtained with perplexity 25 seems to be capturing another pattern as well, since the samples seem to be grouped in pairs or triplets
perplexities = c(2,5,10,25)
ps = list()
for(i in 1:length(perplexities)){
set.seed(123)
tsne = datExpr %>% t %>% Rtsne(perplexity=perplexities[i])
plot_data = data.frame('ID'=colnames(datExpr), 'C1'=tsne$Y[,1], 'C2'=tsne$Y[,2]) %>%
mutate('ID'=substring(ID,2)) %>% left_join(datMeta, by=c('ID'='Dissected_Sample_ID')) %>%
dplyr::select('C1','C2','Diagnosis','Subject_ID') %>%
mutate('Diagnosis'=factor(Diagnosis, levels=c('CTL','ASD')))
ps[[i]] = plot_data %>% ggplot(aes(C1, C2, color=Diagnosis)) + geom_point() + theme_minimal() +
ggtitle(paste0('Perplexity=',perplexities[i])) + theme(legend.position='none')
}
grid.arrange(grobs=ps, nrow=2)
rm(ps, perplexities, tsne, i)The second pattern t-SNE seems to be capturing is the subject the samples belong to!
ggplotly(plot_data %>% ggplot(aes(C1, C2, color=Subject_ID)) + geom_point(aes(id=Subject_ID)) + theme_minimal() +
theme(legend.position='none') + ggtitle('t-SNE Perplexity=20 coloured by Subject ID'))rm(plot_data)Genes
PCA
First Principal Component explains over 99% of the total variance
There’s a really strong correlation between the mean expression of a gene and the 1st principal component
pca = datExpr %>% prcomp
plot_data = data.frame( 'PC1' = pca$x[,1], 'PC2' = pca$x[,2], 'MeanExpr'=rowMeans(datExpr))
plot_data %>% ggplot(aes(PC1, PC2, color=MeanExpr)) + geom_point(alpha=0.3) + theme_minimal() +
scale_color_viridis() + ggtitle('PCA') +
xlab(paste0('PC1 (',round(100*summary(pca)$importance[2,1],1),'%)')) +
ylab(paste0('PC2 (',round(100*summary(pca)$importance[2,2],1),'%)'))
rm(pca, plot_data)MDS
Distance matrix is too heavy to calculate and the resulting distance object is to big to even load (3.4GB)
t-SNE
Higher perplexities capture a cleaner visualisation of the data ordered by mean expression, in a similar (although not as linear) way to PCA
perplexities = c(1,2,5,10,50,100)
ps = list()
for(i in 1:length(perplexities)){
tsne = read.csv(paste0('./../Visualisations/tsne_perplexity_',perplexities[i],'.csv'))
plot_data = data.frame('C1'=tsne[,1], 'C2'=tsne[,2], 'MeanExpr'=rowMeans(datExpr))
ps[[i]] = plot_data %>% ggplot(aes(C1, C2, color=MeanExpr)) + geom_point(alpha=0.5) + theme_minimal() +
scale_color_viridis() + ggtitle(paste0('Perplexity = ',perplexities[i])) + theme(legend.position='none')
}
grid.arrange(grobs=ps, nrow=2)
rm(perplexities, ps, tsne, i)Differential Expression Analysis
4309 genes (~26%) are significant using a threshold of 0.05 for the adjusted p-value and a without a log Fold Change threshold (keeping the null hypothesis \(H_0: lfc=0\))
All genes have an adjusted p-value (there are no NAs)
table(DE_info$padj<0.05, useNA='ifany')##
## FALSE TRUE
## 11642 4505p = DE_info %>% ggplot(aes(log2FoldChange, padj, color=significant)) + geom_point(alpha=0.2) +
scale_y_sqrt() + xlab('log2 Fold Change') + ylab('Adjusted p-value') + theme_minimal()
ggExtra::ggMarginal(p, type = 'density', color='gray', fill='gray', size=10)
rm(p)There is a clear negative relation between lfc and mean expression, for both differentially expressed and not differentially expressed groups of genes
The relation is strongest for genes with low levels of expression
plot_data = data.frame('ID'=rownames(datExpr), 'meanExpr'=rowMeans(datExpr)) %>% left_join(DE_info, by='ID') %>%
mutate('statisticallySignificant'=padj<0.05 & !is.na(padj))
plot_data %>% ggplot(aes(meanExpr, abs(log2FoldChange), color=statisticallySignificant)) + geom_point(alpha=0.1) +
geom_smooth(method='lm', se=FALSE) + theme_minimal() + scale_y_sqrt() + theme(legend.position = 'bottom') +
xlab('Mean Expression') + ylab('abs(lfc)') + ggtitle('Log fold change by level of expression')
When filtering for differential expression, the points separate into two clouds depending on whether they are over or underexpressed
The top cloud corresponds to the over expressed genes and the bottom to the under expressed ones
datExpr_DE = datExpr[DE_info$significant,]
pca = datExpr_DE %>% prcomp
plot_data = cbind(data.frame('PC1'=pca$x[,1], 'PC2'=pca$x[,2]), DE_info[DE_info$significant==TRUE,])
pos_zero = -min(plot_data$log2FoldChange)/(max(plot_data$log2FoldChange)-min(plot_data$log2FoldChange))
p = plot_data %>% ggplot(aes(PC1, PC2, color=log2FoldChange)) + geom_point(alpha=0.5) +
scale_color_gradientn(colours=c('#F8766D','#faa49e','white','#00BFC4','#009499'),
values=c(0, pos_zero-0.05, pos_zero, pos_zero+0.05, 1)) +
theme_minimal() + ggtitle('
PCA of differentially expressed genes') + # This is on purpose, PDF doesn't save well without this white space (?)
xlab(paste0('PC1 (',round(100*summary(pca)$importance[2,1],1),'%)')) +
ylab(paste0('PC2 (',round(100*summary(pca)$importance[2,2],1),'%)')) + theme(legend.position = 'bottom')
ggExtra::ggMarginal(p, type='density', color='gray', fill='gray', size=10)
rm(pos_zero, p)Separating the genes into these two groups: Salmon: under-expressed, aqua: over-expressed
plot_data = plot_data %>% mutate('Group'=ifelse(log2FoldChange>0,'overexpressed','underexpressed')) %>%
mutate('Group' = factor(Group, levels=c('underexpressed','overexpressed')))List of top DE genes
- The genes with the highest LFC are overexpressed genes and most of them don’t has a Neuronal tag
# Get genes names
getinfo = c('ensembl_gene_id','external_gene_id')
mart = useMart(biomart='ENSEMBL_MART_ENSEMBL', dataset='hsapiens_gene_ensembl', host='feb2014.archive.ensembl.org')
gene_names = getBM(attributes=getinfo, filters=c('ensembl_gene_id'), values=plot_data$ID, mart=mart) %>%
rename(external_gene_id = 'gene_name', ensembl_gene_id = 'ID')top_genes = plot_data %>% left_join(gene_names, by='ID') %>% arrange(-abs(log2FoldChange)) %>%
top_n(50, wt=log2FoldChange)
kable(top_genes %>% dplyr::select(ID, gene_name, log2FoldChange, padj, Neuronal, Group))| ID | gene_name | log2FoldChange | padj | Neuronal | Group |
|---|---|---|---|---|---|
| ENSG00000013588 | GPRC5A | 1.9129622 | 0.0000000 | 0 | overexpressed |
| ENSG00000198203 | SULT1C2 | 1.8641403 | 0.0000478 | 0 | overexpressed |
| ENSG00000111057 | KRT18 | 1.8432998 | 0.0000008 | 0 | overexpressed |
| ENSG00000027644 | INSRR | 1.8302265 | 0.0000362 | 0 | overexpressed |
| ENSG00000106366 | SERPINE1 | 1.6476313 | 0.0000002 | 0 | overexpressed |
| ENSG00000164707 | SLC13A4 | 1.6361510 | 0.0001992 | 0 | overexpressed |
| ENSG00000101162 | TUBB1 | 1.5019328 | 0.0005448 | 0 | overexpressed |
| ENSG00000126562 | WNK4 | 1.4832112 | 0.0000029 | 0 | overexpressed |
| ENSG00000124107 | SLPI | 1.4695460 | 0.0000641 | 0 | overexpressed |
| ENSG00000158578 | ALAS2 | 1.4633454 | 0.0013513 | 0 | overexpressed |
| ENSG00000112499 | SLC22A2 | 1.4619922 | 0.0010652 | 0 | overexpressed |
| ENSG00000173110 | HSPA6 | 1.4594581 | 0.0000000 | 0 | overexpressed |
| ENSG00000105641 | SLC5A5 | 1.3904606 | 0.0009910 | 0 | overexpressed |
| ENSG00000186564 | FOXD2 | 1.3887858 | 0.0003363 | 0 | overexpressed |
| ENSG00000108821 | COL1A1 | 1.3887180 | 0.0000047 | 0 | overexpressed |
| ENSG00000064886 | CHI3L2 | 1.3763339 | 0.0001859 | 0 | overexpressed |
| ENSG00000144488 | ESPNL | 1.3722319 | 0.0001190 | 0 | overexpressed |
| ENSG00000176692 | FOXC2 | 1.3576319 | 0.0007596 | 0 | overexpressed |
| ENSG00000168229 | PTGDR | 1.2899250 | 0.0003586 | 0 | overexpressed |
| ENSG00000128342 | LIF | 1.2895419 | 0.0105371 | 1 | overexpressed |
| ENSG00000125735 | TNFSF14 | 1.2801945 | 0.0033952 | 0 | overexpressed |
| ENSG00000099937 | SERPIND1 | 1.2734871 | 0.0014281 | 0 | overexpressed |
| ENSG00000106809 | OGN | 1.2674977 | 0.0005461 | 0 | overexpressed |
| ENSG00000112175 | BMP5 | 1.2591640 | 0.0025202 | 0 | overexpressed |
| ENSG00000170577 | SIX2 | 1.2575209 | 0.0163686 | 0 | overexpressed |
| ENSG00000144644 | GADL1 | 1.2570302 | 0.0084579 | 0 | overexpressed |
| ENSG00000145113 | MUC4 | 1.2417708 | 0.0183299 | 0 | overexpressed |
| ENSG00000171345 | KRT19 | 1.2384708 | 0.0000213 | 0 | overexpressed |
| ENSG00000096696 | DSP | 1.2116499 | 0.0000026 | 0 | overexpressed |
| ENSG00000168542 | COL3A1 | 1.1958974 | 0.0000015 | 1 | overexpressed |
| ENSG00000149452 | SLC22A8 | 1.1782208 | 0.0176148 | 0 | overexpressed |
| ENSG00000126778 | SIX1 | 1.1499890 | 0.0000591 | 1 | overexpressed |
| ENSG00000153404 | PLEKHG4B | 1.1444624 | 0.0000179 | 0 | overexpressed |
| ENSG00000174576 | NPAS4 | 1.1425909 | 0.0044001 | 0 | overexpressed |
| ENSG00000184357 | HIST1H1B | 1.1243519 | 0.0001984 | 0 | overexpressed |
| ENSG00000159212 | CLIC6 | 1.1171812 | 0.0002762 | 0 | overexpressed |
| ENSG00000171819 | ANGPTL7 | 1.0715341 | 0.0063040 | 0 | overexpressed |
| ENSG00000105825 | TFPI2 | 1.0642035 | 0.0034824 | 0 | overexpressed |
| ENSG00000175592 | FOSL1 | 1.0476127 | 0.0089637 | 1 | overexpressed |
| ENSG00000167244 | IGF2 | 1.0462971 | 0.0001178 | 0 | overexpressed |
| ENSG00000269113 | TRABD2B | 1.0456340 | 0.0056013 | 0 | overexpressed |
| ENSG00000173714 | WFIKKN2 | 1.0222116 | 0.0029921 | 0 | overexpressed |
| ENSG00000136542 | GALNT5 | 1.0214133 | 0.0043646 | 0 | overexpressed |
| ENSG00000073792 | IGF2BP2 | 1.0132622 | 0.0000002 | 0 | overexpressed |
| ENSG00000139219 | COL2A1 | 1.0129833 | 0.0195054 | 0 | overexpressed |
| ENSG00000106511 | MEOX2 | 1.0095678 | 0.0000135 | 0 | overexpressed |
| ENSG00000132911 | NMUR2 | 0.9919306 | 0.0037389 | 0 | overexpressed |
| ENSG00000159167 | STC1 | 0.9912194 | 0.0000300 | 0 | overexpressed |
| ENSG00000115648 | MLPH | 0.9760809 | 0.0015824 | 0 | overexpressed |
| ENSG00000173068 | BNC2 | 0.9656989 | 0.0000000 | 0 | overexpressed |
rm(top_genes)Plotting the mean expression by group they seem to have two and three underlying distributions, respectively, so a Gaussian Mixture Model was fitted to each one to separate them into two/three Gaussians and then the points corresponding to each one were plotted in the original PCA plot.
The membership of the genes to the different Gaussians seems to be determined by their level of expression
Under-expressed ASD genes tend to have a higher mean expression than over-expressed ones
gg_colour_hue = function(n) {
hues = seq(15, 375, length = n + 1)
hcl(h = hues, l = 65, c = 100)[1:n]
}
tot_n_clusters = 5
plot_data = plot_data %>% mutate('MeanExpr'=rowMeans(datExpr_DE), 'SDExpr'=apply(datExpr_DE,1,sd))
GMM_G1 = plot_data %>% filter(Group=='overexpressed') %>% dplyr::select(MeanExpr) %>% GMM(3)
GMM_G2 = plot_data %>% filter(Group=='underexpressed') %>% dplyr::select(MeanExpr) %>% GMM(2)
memberships_G1 = data.frame('ID'=plot_data$ID[plot_data$Group=='overexpressed'],
'Membership'=GMM_G1$Log_likelihood %>% apply(1, function(x) glue('over_', which.max(x))))
memberships_G2 = data.frame('ID'=plot_data$ID[plot_data$Group=='underexpressed'],
'Membership'=GMM_G2$Log_likelihood %>% apply(1, function(x) glue('under_', which.max(x))))
plot_data = rbind(memberships_G1, memberships_G2) %>% left_join(plot_data, by='ID')
p1 = plot_data %>% ggplot(aes(x=MeanExpr, color=Group, fill=Group)) + geom_density(alpha=0.4) +
theme_minimal() + theme(legend.position='bottom')
p2 = plot_data %>% ggplot(aes(x=MeanExpr)) +
stat_function(fun=dnorm, n=100, colour=gg_colour_hue(tot_n_clusters)[1],
args=list(mean=GMM_G1$centroids[1], sd=GMM_G1$covariance_matrices[1])) +
stat_function(fun=dnorm, n=100, colour=gg_colour_hue(tot_n_clusters)[2],
args=list(mean=GMM_G1$centroids[2], sd=GMM_G1$covariance_matrices[2])) +
stat_function(fun=dnorm, n=100, colour=gg_colour_hue(tot_n_clusters)[3],
args=list(mean=GMM_G1$centroids[3], sd=GMM_G1$covariance_matrices[3])) +
stat_function(fun=dnorm, n=100, colour=gg_colour_hue(tot_n_clusters)[4],
args=list(mean=GMM_G2$centroids[1], sd=GMM_G2$covariance_matrices[1])) +
stat_function(fun=dnorm, n=100, colour=gg_colour_hue(tot_n_clusters)[5],
args=list(mean=GMM_G2$centroids[2], sd=GMM_G2$covariance_matrices[2])) +
theme_minimal()
p3 = plot_data %>% ggplot(aes(PC1, PC2, color=Membership)) + geom_point(alpha=0.4) + theme_minimal() +
theme(legend.position='bottom')
grid.arrange(p1, p2, p3, nrow=1)
rm(gg_color_hue, n_clusters, GMM_G1, GMM_G2, memberships_G1, memberships_G2, p1, p2, p3, tot_n_clusters)For previous preprocessing pipelines, the pattern found above was also present in the standard deviation, but there doesn’t seem to be any distinguishable patterns now. This could be because the variance was almost homogenised with the vst normalisation algorithm.
plot_data %>% ggplot(aes(x=SDExpr, color=Group, fill=Group)) + geom_density(alpha=0.4) + theme_minimal()
rm(plot_data)Effects of modifying the log fold change threshold
fc_list = seq(1, 1.25, 0.01)
n_genes = nrow(datExpr)
# Calculate PCAs
datExpr_pca_samps = datExpr %>% data.frame %>% t %>% prcomp(scale.=TRUE)
datExpr_pca_genes = datExpr %>% data.frame %>% prcomp(scale.=TRUE)
# Initialise DF to save PCA outputs
pcas_samps = datExpr_pca_samps$x %>% data.frame %>% dplyr::select(PC1:PC2) %>%
mutate('ID'=colnames(datExpr), 'fc'=-1, PC1=scale(PC1), PC2=scale(PC2))
pcas_genes = datExpr_pca_genes$x %>% data.frame %>% dplyr::select(PC1:PC2) %>%
mutate('ID'=rownames(datExpr), 'fc'=-1, PC1=scale(PC1), PC2=scale(PC2))
pca_samps_old = pcas_samps
pca_genes_old = pcas_genes
for(fc in fc_list){
# Recalculate DE_info with the new threshold (p-values change) an filter DE genes
DE_genes = results(dds, lfcThreshold=log2(fc), altHypothesis='greaterAbs') %>% data.frame %>%
mutate('ID'=rownames(.)) %>% filter(padj<0.05)
datExpr_DE = datExpr %>% data.frame %>% filter(rownames(.) %in% DE_genes$ID)
n_genes = c(n_genes, nrow(DE_genes))
# Calculate PCAs
datExpr_pca_samps = datExpr_DE %>% t %>% prcomp(scale.=TRUE)
datExpr_pca_genes = datExpr_DE %>% prcomp(scale.=TRUE)
# Create new DF entries
pca_samps_new = datExpr_pca_samps$x %>% data.frame %>% dplyr::select(PC1:PC2) %>%
mutate('ID'=colnames(datExpr), 'fc'=fc, PC1=scale(PC1), PC2=scale(PC2))
pca_genes_new = datExpr_pca_genes$x %>% data.frame %>% dplyr::select(PC1:PC2) %>%
mutate('ID'=DE_genes$ID, 'fc'=fc, PC1=scale(PC1), PC2=scale(PC2))
# Change PC sign if necessary
if(cor(pca_samps_new$PC1, pca_samps_old$PC1)<0) pca_samps_new$PC1 = -pca_samps_new$PC1
if(cor(pca_samps_new$PC2, pca_samps_old$PC2)<0) pca_samps_new$PC2 = -pca_samps_new$PC2
if(cor(pca_genes_new[pca_genes_new$ID %in% pca_genes_old$ID,]$PC1,
pca_genes_old[pca_genes_old$ID %in% pca_genes_new$ID,]$PC1)<0){
pca_genes_new$PC1 = -pca_genes_new$PC1
}
if(cor(pca_genes_new[pca_genes_new$ID %in% pca_genes_old$ID,]$PC2,
pca_genes_old[pca_genes_old$ID %in% pca_genes_new$ID,]$PC2 )<0){
pca_genes_new$PC2 = -pca_genes_new$PC2
}
pca_samps_old = pca_samps_new
pca_genes_old = pca_genes_new
# Update DFs
pcas_samps = rbind(pcas_samps, pca_samps_new)
pcas_genes = rbind(pcas_genes, pca_genes_new)
}
# Add Diagnosis/SFARI score information
pcas_samps = pcas_samps %>% mutate('ID'=substring(ID,2)) %>%
left_join(datMeta, by=c('ID'='Dissected_Sample_ID')) %>%
dplyr::select(ID, PC1, PC2, fc, Diagnosis, Brain_lobe)
# pcas_genes = pcas_genes %>% left_join(SFARI_genes, by='ID') %>%
# mutate('score'=as.factor(`gene-score`)) %>%
# dplyr::select(ID, PC1, PC2, lfc, score)
# Plot change of number of genes
ggplotly(data.frame('lfc'=log2(fc_list), 'n_genes'=n_genes[-1]) %>% ggplot(aes(x=lfc, y=n_genes)) +
geom_point() + geom_line() + theme_minimal() + xlab('|Log Fold Change|') + ylab('Remaining Genes') +
ggtitle('Remaining genes when modifying filtering threshold'))rm(fc_list, n_genes, fc, pca_samps_new, pca_genes_new, pca_samps_old, pca_genes_old,
datExpr_pca_samps, datExpr_pca_genes)Samples
Note: PC values get smaller as Log2 fold change increases, so on each iteration the values were scaled so it would be easier to compare between frames
Coloured by Diagnosis:
LFC = -1 represents the whole set of genes, without any filtering by differential expression
Log Fold Changes between 0.06 and 0.18 seem to separate the samples by Diagnosis best
ggplotly(pcas_samps %>% mutate(abs_lfc=ifelse(fc==-1,-1,round(log2(fc),2))) %>%
ggplot(aes(PC1, PC2, color=Diagnosis)) + geom_point(aes(frame=abs_lfc, ids=ID), alpha=0.7) +
theme_minimal() + ggtitle('Samples PCA plot modifying filtering threshold'))Coloured by brain region:
There doesn’t seem to be any recognisable pattern
ggplotly(pcas_samps %>% mutate(abs_lfc=ifelse(fc==-1,-1,round(log2(fc),2))) %>%
ggplot(aes(PC1, PC2, color=Brain_lobe)) + geom_point(aes(frame=abs_lfc, ids=ID)) +
theme_minimal() + ggtitle('Samples PCA plot modifying filtering threshold'))Genes
if(!'fcSign' %in% colnames(pcas_genes)){
pcas_genes = pcas_genes %>% left_join(DE_info, by='ID') %>% mutate(fcSign = ifelse(log2FoldChange>0,'Positive','Negative'))
}
ggplotly(pcas_genes %>% mutate(abs_lfc=ifelse(fc==-1,-1,round(log2(fc),2))) %>%
ggplot(aes(PC1, PC2, color=fcSign)) + geom_point(aes(frame=abs_lfc, ids=ID), alpha=0.2) +
theme_minimal() + ggtitle('Genes PCA plot modifying |LFC| filtering threshold'))